RuntimeError: Expected object of scalar type Long but got scalar type Float for sequence element 1 i

报错

Traceback (most recent call last):
  File "main.py", line 306, in <module>
    args_config=args_config,
  File "main.py", line 224, in train
    avg_reward,
  File "main.py", line 56, in train_one_epoch
    selected_neg_items_list, _ = sampler(batch_data, adj_matrix, edge_matrix)
  File "/home/zzy/anaconda3/lib/python3.7/site-packages/torch/nn/modules/module.py", line 493, in __call__
    result = self.forward(*input, **kwargs)
  File "/home/zzy/KG-Policy/kgpolicy/modules/sampler/kgpolicy.py", line 81, in forward
    self.edges = self.build_edge(edge_matrix)
  File "/home/zzy/KG-Policy/kgpolicy/modules/sampler/kgpolicy.py", line 122, in build_edge
    edges = torch.cat((node_index.unsqueeze(1), neighbor_index.unsqueeze(1)), dim=1)
RuntimeError: Expected object of scalar type Long but got scalar type Float for sequence element 1 in sequence argument at position #1 'tensors'

在报错的地方:

RuntimeError: Expected object of scalar type Long but got scalar type Float for sequence element 1 in sequence argument at position #1 ‘tensors’

显然是类型错误,应该是long,但是实际是float
涉及到的文件是main.py和kgpolicy.py
找到kgpolicy.py的122行
发现是torch.cat()函数里面的参数问题
所以改为:

edges = torch.cat((node_index.unsqueeze(1).long(), neighbor_index.unsqueeze(1).long()), dim=1) #已修改

即可.当时找了很久,一直不知道应该加在哪里.就按照错误提示行,一个个看.有函数的就到具体的函数位置寻找可能是错误的地方就行了.